L Erickson

Spatial Analysis of Residential Mortgage Default in a Metropolitan County

Abstract In this pilot study, research was conducted to explore the application of GIS technology to residential mortgage default. The purpose of this study was to examine the relationship between foreclosure rates and neighborhood characteristics in a metropolitan county in Utah. Neighborhood was defined as a discrete spatial entity that contains households and housing structures with similar characteristics, and is equivalent to a census tract. Regression analysis indicated that characteristics such as proportion of household units with second mortgages and racial composition of census tracts explained 80% of the variance of foreclosure rates among the communities. The geo-mapping analysis of Cache County showed that foreclosure rates were spatially differentiated. Clusters of foreclosures were prominent in the southwestern portion of the county.

Concentrations of Subprime LendingActivities, Equity Risk Factors, and Bankruptcy Filings

Abstract High levels of bankruptcy filings in Utah along with high concentrations of subprime originated and refinanced mortgage loans raised the question of whether or not these two indicators were related. Aggregate level studies of subprime market activities have suggested the spatial nature of lending, but they have not tested it in relation to bankruptcy filings. Using multivariate analysis it was shown that the aggregated subprime market activity and equity-related risk factors, such as the use of second mortgages and estimations of the risk-related investments in real estate at the local level, had important and statistically significant effects on aggregate bankruptcy filings. Results from the model suggested that about 32% of the variance in Utah bankruptcy filings in 2003 was accounted for by housing-related predictors.